Current Research Interest
My main research interest has been large-scale inference and
massive-data analysis, where the data are usually very high-dimensional
and one must estimate very large numbers of parameters or test very large
numbers of hypotheses simultaneously. The setting is frequently found
in many scientific areas, e.g. genomics, astronomy, functional
Magnetic Resonance Imaging (fMRI), and image processing.
Advances in large-scale inferences
enable faster exactration of useful information in various
scientific fieds and broaden the scope of theory and methodology in
statistics.
Theory
- Larg-Scale Multiple Hypotheses Testing: inference on the proportion
of non-null effects, inference on null parameters, False
Discovery Rate (FDR)-controlling methodology.
- High Dimensional inference: simultaneous
estimation and variable selection by exploiting sparsity, basis pursuit
through L^1
penalization, uncertainty principle.
- Others: Decision Theory and Asymptotics, Random Matrix Theory (RMT),
Wavelets
and Signal Processing.
Applications
- Astronomy and Cosmology: non-Gaussian signature detection in
Cosmic Micorwave Background (CMB).
- Genomics: gene microarray,
Comparative Genomic Hybridization (CGH), protein mass spectroscopy.
- Others: computer security, watermarking.
-
Link
to Astrostatistics group at Purdue
Reports
Click here to see
reports listed by different areas
2006
2005
Talks